In gabrielburcea/cvindia: Analysis of symptoms

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  echo = FALSE,
  warning = FALSE,
  message = FALSE
)

library(magrittr)

Several approaches were taken in order to achieve the following analysis.

I. First batch of analysis is descriptive in nature. It is an attempt in Covid-19 symptom mapping. II. With the first step findings achieved, second step focuses on whether there are differences in Covid-19 symptom in different comorbidity groups of respondents. Univariate analysis and multivariate logistic regression has aid the second stage analysis.

data_select <- readr::read_csv("/Users/gabrielburcea/Rprojects/data/data_descr_15062020.csv")

data_select %>% dplyr::distinct(covid_tested)
#data_map <- readr::read_csv("../Rprojects/data/PivotMappe060520r.csv")

# world_map <- cvindia::count_mapp_world(data = data_select, start_date = as.Date("2020-04-09", format = "%Y-%m-%d"),
#                          end_date = as.Date("2020-05-09", format = "%Y-%m-%d" ), plot_chart = TRUE)
# 
# world_map

# world_map <- cvindia::count_mapp_world(data = data_map, start_date = as.Date("2020-04-09", format = "%Y-%m-%d"),
#                          end_date = as.Date("2020-05-09", format = "%Y-%m-%d" ), plot_chart = FALSE)
# 
# knitr::kable(world_map)

I. Descriptive analytics

Exploratory questions of interest are: 1. How does symptom profile map to mild/moderate/severe? 2. Does symptom profile differ by underlying co-morbidity?

Answering to first question: 1. How does symptom profile map to mild/moderate/severe?

We choose to look at the most common symptoms according to UK guidelines in patients who reported that they had a Covid 19 positive test. We look at these symptoms according to whether they were reported as mild/moderate/severe

symtoms_covid_in_resp <- cvindia::symptom_profile_covid_tested(data = data_select, start_date = as.Date("2020-04-01", format = "%Y-%m-%d"), 
                                           end_date = as.Date("2020-09-01", format = "%Y-%m-%d"), plot_chart = TRUE)
symtoms_covid_in_resp

Cough, followed by muscle ache are the most symptoms in patient who declared they had a Covid-19 positive test. Loss of smell and shorthness of breath are next. Yet, as observed there are different levels of severity, were mild form of symptoms are reported.

symtoms_covid_in_resp <- cvindia::symptom_profile_covid_tested(data = data_select, start_date = as.Date("2020-04-01", format = "%Y-%m-%d"), 
                                           end_date = as.Date("2020-09-01", format = "%Y-%m-%d"), plot_chart = FALSE)

sympt_cov_in_resp_n <- dplyr::top_n(symtoms_covid_in_resp, 10)

knitr::kable(dplyr::top_n(sympt_cov_in_resp_n, 10))

The next barchart takes into accound all other symptoms for the purpose of comparison. However, fatigue seems to have the highest countS, followed by the most common symptoms of Covid-19.

all_syptpms_frequency <- cvindia::symptom_profile_frequency_plot(data = data_select, start_date = as.Date("2020-04-09", format = "%Y-%m-%d"), 
                                            end_date = as.Date("2020-05-09", format = "%Y-%m-%d"),
                                            plot_chart = TRUE)

all_syptpms_frequency

Table bellow shows the first ten most occurring symptoms taking into account the level of severity. However, the most common symptoms expressed in level of severity are Cough as a mild for, shortness of breath, fatigue, muscle ache and headache, all in a mild form.

all_syptpms_numbers <- cvindia::symptom_profile_frequency_plot(data = data_select, start_date = as.Date("2020-04-09", format = "%Y-%m-%d"), 
                                            end_date = as.Date("2020-05-09", format = "%Y-%m-%d"),
                                            plot_chart = FALSE)

all_sympt_numb <- dplyr::top_n(all_syptpms_numbers, 10)


ten_most_occuring_symptoms <- dplyr::top_n(all_sympt_numb, 10)

knitr::kable(ten_most_occuring_symptoms)

Most people who are completing are reporting their symptoms as mild. However, this are non-generalizable cohort.

diagn_mild_mod_severe <- cvindia::self_diagnosis_mild_mod_severe(data = data_select, start_date = as.Date("2020-04-09"), end_date = as.Date("2020-05-09"), plot_chart = TRUE, title = "Self Report")

diagn_mild_mod_severe

self_reported <- cvindia::self_diagnosis_mild_mod_severe(data = data_select, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = FALSE, title = "Test")


self_reported_n <- dplyr::top_n(self_reported, 10)

knitr::kable(self_reported_n)

Respondents who tested covid positive have reported the 37.5-38 temperature as the most frequent.

temperature <-  cvindia::temparature_covid_tested(data = data_sel, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = TRUE)

temperature

temperature <-  cvindia::temparature_covid_tested(data = data_select, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = FALSE)

temperature

The bellow barchart maps the patients who claimed their symptoms were overall mild. What symptoms did they have? The most prevalent symptoms are cough, fatigue, sore throat, sputum and muscle ache.

mild <- cvindia::symptom_profile_mild(data = data_select, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = TRUE)

mild

The most prevalent symptoms are cough, fatigue, sore throat, sputum and muscle ache.

mild <- cvindia::symptom_profile_mild(data = data_select, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = FALSE)

knitr::kable(dplyr::top_n(mild, 10))

Bellow chart is mapping the patients claiming their symptoms were overall moderate. What symptoms did they have?

moderate <- cvindia::moderate_manifestation_plot(data = data_select, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = TRUE)

moderate

Shortness of breath, cough, sputum fatigue as a mild form followed by moderate fatigue are the first 5 symptoms declared by the respondents of Your.md app.

moderate <- cvindia::moderate_manifestation_plot(data = data_select, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = FALSE)

moderate_n <- dplyr::top_n(moderate, 10)
knitr::kable(moderate_n)

Bellow chart is mapping the patients claiming their symptoms were overall severe, What symptoms did they have?

severe <- cvindia::severe_manifestation_plot(data = data_select, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = TRUE)

severe

Cough, sputum, sore throat, shortness of breath and headache in mild form are the first 5 symptoms.

severe <- cvindia::severe_manifestation_plot(data = data_select, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = FALSE)

severe_n <- dplyr::top_n(severe, 10)

knitr::kable(severe_n)

As a conclusion we may pressume a symptom trajectory in covid-19 positive tested. Mild sore throat which then progresses to a cough, and shortness of breath?

Does symptom profile differ by underlying co-morbidity?

The bar chart bellow shows symptom across co-morbidity groups. By observing the obesity and hypertensive patients, although we see a similar pattern in symptom manifestation, there are slight differences when it comes to sputum and sore throat. In hypertensive respondents sore throat is more prominent than in obese respondents, which are experiencing more sputum. This is not the same with respondent with asthma, which are experiencing sputum and shortness of breath, where sore throat comes on the fourth place, after muscle ache.

data_select <- read_csv("/Users/gabrielburcea/Rprojects/data/PivotMappe060520r.csv")


comorb_cov_sympt <- cvindia::comorbidities_symptoms(data = data_select, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = TRUE)


comorb_cov_sympt

no_cov <- function(data,start_date = as.Date("2020-04-09", tz = "Europe/London"),
                              end_date = as.Date("2020-05-06",tz = "Europe/London"), plot_chart = TRUE){


  show_sympt_cov <- data %>%
    dplyr::select(id, covid_tested, chills, cough, diarrhoea, fatigue, headache, loss_smell_taste, muscle_ache, nasal_congestion, nausea_vomiting, 
                  shortness_breath, sore_throat, sputum, temperature) %>%
    tidyr::pivot_longer(cols=3:15, names_to="Symptom", values_to="Answer") %>%
    dplyr::filter(covid_tested == 'none') %>%
    dplyr::group_by(Symptom, Answer) %>%
    dplyr::summarise(Count = n()) %>%
    dplyr::mutate(Percent = Count/sum(Count)) %>%
    dplyr::filter(Symptom != 'temperature')


  title_stub <- "Symptoms in respondents with no covid\n"
  start_date_title <- format(as.Date(start_date), format = "%d %B %Y")
  end_date_title <- format(as.Date(end_date), format = "%d %B %Y")
  chart_title <- paste0(title_stub, start_date_title, " to ", end_date_title)


  plot_show_symptoms <-
    ggplot2::ggplot(show_sympt_cov, ggplot2::aes(x = reorder(Symptom,-Count), Count, fill = Answer)) +
    ggplot2::coord_flip() +
    ggplot2::geom_bar(stat = "identity", position = "dodge") +
    ggplot2::scale_x_discrete(limits = unique(show_sympt_cov$Symptom)) +
    ggplot2::theme(legend.position = "bottom") +
    ggplot2::scale_fill_brewer(palette="Oranges")+
    ggplot2::theme_minimal() +
    ggplot2::guides(fill = ggplot2::guide_legend(nrow = 3)) +
    ggplot2::labs( title = chart_title,
                   subtitle = "Counts of symptoms in respondents with no Covid-19",
                   y = "Counts",
                   x = "Symptoms",
                   caption = "Source: Dataset - Your.md Dataset") +
    ggplot2::theme(
      axis.title.y = ggplot2::element_text(margin = ggplot2::margin(
        t = 0,
        r = 21,
        b = 0,
        l = 0
      )),
      plot.title = ggplot2::element_text(size = 10, face = "bold"),
      plot.subtitle = ggplot2::element_text(size = 9),
      legend.position = "bottom",
      legend.box = "horizontal",
      axis.text.x = ggplot2::element_text(angle = 55, hjust = 1)
    )

  plot_show_symptoms

  if(plot_chart == TRUE){


    plot_show_symptoms

  }else{

    show_sympt_cov

  }
}

library(tidyverse)


cov_sympt_plot <- cvindia::no_cov(data = data_no_sev,start_date = as.Date("2020-04-09", tz = "Europe/London"),
                              end_date = as.Date("2020-05-06",tz = "Europe/London"), plot_chart = TRUE)

plotly::ggplotly(cov_sympt_plot)

comorb_cov_sympt_num <- cvindia::comorbidities_symptoms(data = data_select, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = FALSE)


comorb_cov_sympt_num


#knitr::kable(dplyr::top_n(comorb_cov_sympt_num,10))

hypertension_num <- comorb_cov_sympt_num %>%
  dplyr::filter(Morbidity == "High Blood Pressure (hypertension)") %>%
  dplyr::arrange(desc(Count))


start_date <- as.Date("2020-04-09")
end_date <- as.Date("2020-05-09")

title_stub <- "Hypertension across symptoms\n"
start_date_title <- format(as.Date(start_date), format = "%d %B %Y")
end_date_title <- format(as.Date(end_date), format = "%d %B %Y")
chart_title <- paste0(title_stub, start_date_title, " to ", end_date_title)



plot_hypertension_sympt <-
    ggplot2::ggplot(hypertension_num, ggplot2::aes(x = Symptom, y = Count, fill = Count)) +
    ggplot2::coord_flip() +
    ggplot2::geom_bar(stat = "identity", position = "dodge") +
    ggplot2::scale_fill_viridis_c(option = "magma", direction = -1) +
    ggplot2::scale_x_discrete(limits = unique(hypertension_num$Symptom)) +
    #ggplot2::theme(legend.position = "bottom") +
    #ggplot2::guides(fill = ggplot2::guide_legend(nrow = 3)) +
    ggplot2::theme_minimal() +
    ggplot2::labs( title = chart_title,
                   subtitle = "Counts of patients with comorbidities accross symptoms",
                   y = "Counts",
                   x = "Symptoms",
                   caption = "Source: Dataset - Your.md Dataset") +
    ggplot2::theme(
      axis.title.y = ggplot2::element_text(margin = ggplot2::margin(
        t = 0,
        r = 21,
        b = 0,
        l = 0
      )),
      plot.title = ggplot2::element_text(size = 10, face = "bold"),
      plot.subtitle = ggplot2::element_text(size = 9),
      axis.text.x = ggplot2::element_text(angle = 55, hjust = 1)
    )


plotly::ggplotly(plot_hypertension_sympt)

obesity_num <- comorb_cov_sympt_num %>%
  dplyr::filter(Morbidity == "Obesity") %>%
  dplyr::arrange(desc(Count))


title_stub <- "Obesity across symptoms\n"
start_date_title <- format(as.Date(start_date), format = "%d %B %Y")
end_date_title <- format(as.Date(end_date), format = "%d %B %Y")
chart_title <- paste0(title_stub, start_date_title, " to ", end_date_title)


plot_obesity_sympt <-
    ggplot2::ggplot(obesity_num, ggplot2::aes(x = Symptom, y = Count, fill = Count)) +
    ggplot2::coord_flip() +
    ggplot2::geom_bar(stat = "identity", position = "dodge") +
    ggplot2::scale_fill_viridis_c(option = "magma", direction = -1) +
    ggplot2::scale_x_discrete(limits = unique(obesity_num$Symptom)) +
    #ggplot2::theme(legend.position = "bottom") +
    #ggplot2::guides(fill = ggplot2::guide_legend(nrow = 3)) +
    ggplot2::theme_minimal() +
    ggplot2::labs( title = chart_title,
                   subtitle = "Counts of patients with comorbidities accross symptoms",
                   y = "Counts",
                   x = "Symptoms",
                   caption = "Source: Dataset - Your.md Dataset") +
    ggplot2::theme(
      axis.title.y = ggplot2::element_text(margin = ggplot2::margin(
        t = 0,
        r = 21,
        b = 0,
        l = 0
      )),
      plot.title = ggplot2::element_text(size = 10, face = "bold"),
      plot.subtitle = ggplot2::element_text(size = 9),
      axis.text.x = ggplot2::element_text(angle = 55, hjust = 1)
    )


plotly::ggplotly(plot_obesity_sympt)

asthma_num <- comorb_cov_sympt_num %>%
  dplyr::filter(Morbidity == "Asthma (managed with an inhaler)") %>%
  dplyr::arrange(desc(Count))


title_stub <- "Asthma across symptoms\n"
start_date_title <- format(as.Date(start_date), format = "%d %B %Y")
end_date_title <- format(as.Date(end_date), format = "%d %B %Y")
chart_title <- paste0(title_stub, start_date_title, " to ", end_date_title)


plot_asthma_sympt <-
   ggplot2::ggplot(asthma_num, ggplot2::aes(x = Symptom, y = Count, fill = Count)) +
    ggplot2::coord_flip() +
    ggplot2::geom_bar(stat = "identity", position = "dodge") +
    ggplot2::scale_fill_viridis_c(option = "magma", direction = -1) +
    ggplot2::scale_x_discrete(limits = unique(asthma_num$Symptom)) +
    #ggplot2::theme(legend.position = "bottom") +
    #ggplot2::guides(fill = ggplot2::guide_legend(nrow = 3)) +
    ggplot2::theme_minimal() +
    ggplot2::labs( title = chart_title,
                   subtitle = "Counts of patients with comorbidities accross symptoms",
                   y = "Counts",
                   x = "Symptoms",
                   caption = "Source: Dataset - Your.md Dataset") +
    ggplot2::theme(
      axis.title.y = ggplot2::element_text(margin = ggplot2::margin(
        t = 0,
        r = 21,
        b = 0,
        l = 0
      )),
      plot.title = ggplot2::element_text(size = 10, face = "bold"),
      plot.subtitle = ggplot2::element_text(size = 9),
      axis.text.x = ggplot2::element_text(angle = 55, hjust = 1)
    )


plotly::ggplotly(plot_asthma_sympt)

lung_num <- comorb_cov_sympt_num %>%
  dplyr::filter(Morbidity == "Long-Standing Lung Condition") %>%
  dplyr::arrange(desc(Count))


title_stub <- "Long-Standing Lung Condition across symptoms\n"
start_date_title <- format(as.Date(start_date), format = "%d %B %Y")
end_date_title <- format(as.Date(end_date), format = "%d %B %Y")
chart_title <- paste0(title_stub, start_date_title, " to ", end_date_title)


plot_lung_sympt <-
    ggplot2::ggplot(lung_num, ggplot2::aes(x = Symptom, y = Count, fill = Count)) +
    ggplot2::coord_flip() +
    ggplot2::geom_bar(stat = "identity", position = "dodge") +
    ggplot2::scale_fill_viridis_c(option = "magma", direction = -1) +
    ggplot2::scale_x_discrete(limits = unique(lung_num$Symptom)) +
    #ggplot2::theme(legend.position = "bottom") +
    #ggplot2::guides(fill = ggplot2::guide_legend(nrow = 3)) +
    ggplot2::theme_minimal() +
    ggplot2::labs( title = chart_title,
                   subtitle = "Counts of patients with comorbidities accross symptoms",
                   y = "Counts",
                   x = "Symptoms",
                   caption = "Source: Dataset - Your.md Dataset") +
    ggplot2::theme(
      axis.title.y = ggplot2::element_text(margin = ggplot2::margin(
        t = 0,
        r = 21,
        b = 0,
        l = 0
      )),
      plot.title = ggplot2::element_text(size = 10, face = "bold"),
      plot.subtitle = ggplot2::element_text(size = 9),
      axis.text.x = ggplot2::element_text(angle = 55, hjust = 1)
    )


plotly::ggplotly(plot_lung_sympt)

heart_num <- comorb_cov_sympt_num %>%
  dplyr::filter(Morbidity == "Long-Standing Heart Disease") %>%
  dplyr::arrange(desc(Count))


title_stub <- "Long-Standing Heart Disease across symptoms\n"
start_date_title <- format(as.Date(start_date), format = "%d %B %Y")
end_date_title <- format(as.Date(end_date), format = "%d %B %Y")
chart_title <- paste0(title_stub, start_date_title, " to ", end_date_title)


plot_heart_sympt <-
    ggplot2::ggplot(heart_num, ggplot2::aes(x = Symptom, y = Count, fill = Count)) +
    ggplot2::coord_flip() +
    ggplot2::geom_bar(stat = "identity", position = "dodge") +
    ggplot2::scale_fill_viridis_c(option = "magma", direction = -1) +
    ggplot2::scale_x_discrete(limits = unique(heart_num$Symptom)) +
    #ggplot2::theme(legend.position = "bottom") +
    #ggplot2::guides(fill = ggplot2::guide_legend(nrow = 3)) +
    ggplot2::theme_minimal() +
    ggplot2::labs( title = chart_title,
                   subtitle = "Counts of patients with comorbidities accross symptoms",
                   y = "Counts",
                   x = "Symptoms",
                   caption = "Source: Dataset - Your.md Dataset") +
    ggplot2::theme(
      axis.title.y = ggplot2::element_text(margin = ggplot2::margin(
        t = 0,
        r = 21,
        b = 0,
        l = 0
      )),
      plot.title = ggplot2::element_text(size = 10, face = "bold"),
      plot.subtitle = ggplot2::element_text(size = 9),
      axis.text.x = ggplot2::element_text(angle = 55, hjust = 1)
    )


plotly::ggplotly(plot_heart_sympt)

kidney_num <- comorb_cov_sympt_num %>%
  dplyr::filter(Morbidity == "Long-Standing Kidney Disease") %>%
  dplyr::arrange(desc(Count))


title_stub <- "Long-Standing Kidney Disease across symptoms\n"
start_date_title <- format(as.Date(start_date), format = "%d %B %Y")
end_date_title <- format(as.Date(end_date), format = "%d %B %Y")
chart_title <- paste0(title_stub, start_date_title, " to ", end_date_title)


plot_kidney_sympt <-
    ggplot2::ggplot(kidney_num, ggplot2::aes(x = Symptom, y = Count, fill = Count)) +
    ggplot2::coord_flip() +
    ggplot2::geom_bar(stat = "identity", position = "dodge") +
    ggplot2::scale_fill_viridis_c(option = "magma", direction = -1) +
    ggplot2::scale_x_discrete(limits = unique(kidney_num$Symptom)) +
    #ggplot2::theme(legend.position = "bottom") +
    #ggplot2::guides(fill = ggplot2::guide_legend(nrow = 3)) +
    ggplot2::theme_minimal() +
    ggplot2::labs( title = chart_title,
                   subtitle = "Counts of patients with comorbidities accross symptoms",
                   y = "Counts",
                   x = "Symptoms",
                   caption = "Source: Dataset - Your.md Dataset") +
    ggplot2::theme(
      axis.title.y = ggplot2::element_text(margin = ggplot2::margin(
        t = 0,
        r = 21,
        b = 0,
        l = 0
      )),
      plot.title = ggplot2::element_text(size = 10, face = "bold"),
      plot.subtitle = ggplot2::element_text(size = 9),
      axis.text.x = ggplot2::element_text(angle = 55, hjust = 1)
    )


plotly::ggplotly(plot_kidney_sympt)

liver_num <- comorb_cov_sympt_num %>%
  dplyr::filter(Morbidity == "Long-Standing Liver Disease") %>%
  dplyr::arrange(desc(Count))


title_stub <- "Long-Standing Liver Disease across symptoms\n"
start_date_title <- format(as.Date(start_date), format = "%d %B %Y")
end_date_title <- format(as.Date(end_date), format = "%d %B %Y")
chart_title <- paste0(title_stub, start_date_title, " to ", end_date_title)


plot_liver_num <-
    ggplot2::ggplot(liver_num, ggplot2::aes(x = Symptom, y = Count, fill = Count)) +
    ggplot2::coord_flip() +
    ggplot2::geom_bar(stat = "identity", position = "dodge") +
    ggplot2::scale_fill_viridis_c(option = "magma", direction = -1) +
    ggplot2::scale_x_discrete(limits = unique(liver_num$Symptom)) +
    #ggplot2::theme(legend.position = "bottom") +
    #ggplot2::guides(fill = ggplot2::guide_legend(nrow = 3)) +
    ggplot2::theme_minimal() +
    ggplot2::labs( title = chart_title,
                   subtitle = "Counts of patients with comorbidities accross symptoms",
                   y = "Counts",
                   x = "Symptoms",
                   caption = "Source: Dataset - Your.md Dataset") +
    ggplot2::theme(
      axis.title.y = ggplot2::element_text(margin = ggplot2::margin(
        t = 0,
        r = 21,
        b = 0,
        l = 0
      )),
      plot.title = ggplot2::element_text(size = 10, face = "bold"),
      plot.subtitle = ggplot2::element_text(size = 9),
      axis.text.x = ggplot2::element_text(angle = 55, hjust = 1)
    )


plotly::ggplotly(plot_liver_num)

diab_type_one_num <- comorb_cov_sympt_num %>%
  dplyr::filter(Morbidity == "Diabetes Type 1 (controlled by insulin)") %>%
  dplyr::arrange(desc(Count))


title_stub <- "Diabetes Type 1 across symptoms\n"
start_date_title <- format(as.Date(start_date), format = "%d %B %Y")
end_date_title <- format(as.Date(end_date), format = "%d %B %Y")
chart_title <- paste0(title_stub, start_date_title, " to ", end_date_title)


plot_diab_1_num <-
    ggplot2::ggplot(diab_type_one_num, ggplot2::aes(x = Symptom, y = Count, fill = Count)) +
    ggplot2::coord_flip() +
    ggplot2::geom_bar(stat = "identity", position = "dodge") +
    ggplot2::scale_fill_viridis_c(option = "magma", direction = -1) +
    ggplot2::scale_x_discrete(limits = unique(diab_type_one_num$Symptom)) +
    #ggplot2::theme(legend.position = "bottom") +
    #ggplot2::guides(fill = ggplot2::guide_legend(nrow = 3)) +
    ggplot2::theme_minimal() +
    ggplot2::labs( title = chart_title,
                   subtitle = "Counts of patients with comorbidities accross symptoms",
                   y = "Counts",
                   x = "Symptoms",
                   caption = "Source: Dataset - Your.md Dataset") +
    ggplot2::theme(
      axis.title.y = ggplot2::element_text(margin = ggplot2::margin(
        t = 0,
        r = 21,
        b = 0,
        l = 0
      )),
      plot.title = ggplot2::element_text(size = 10, face = "bold"),
      plot.subtitle = ggplot2::element_text(size = 9),
      axis.text.x = ggplot2::element_text(angle = 55, hjust = 1)
    )


plotly::ggplotly(plot_diab_1_num)

diab_type_two_num <- comorb_cov_sympt_num %>%
  dplyr::filter(Morbidity == "Diabetes Type 2") %>%
  dplyr::arrange(desc(Count))


title_stub <- "Diabetes Type 2 across symptoms\n"
start_date_title <- format(as.Date(start_date), format = "%d %B %Y")
end_date_title <- format(as.Date(end_date), format = "%d %B %Y")
chart_title <- paste0(title_stub, start_date_title, " to ", end_date_title)


plot_diab_2_num <-
    ggplot2::ggplot(diab_type_two_num, ggplot2::aes(x = Symptom, y = Count, fill = Count)) +
    ggplot2::coord_flip() +
    ggplot2::geom_bar(stat = "identity", position = "dodge") +
    ggplot2::scale_fill_viridis_c(option = "magma", direction = -1) +
    ggplot2::scale_x_discrete(limits = unique(diab_type_two_num$Symptom)) +
    #ggplot2::theme(legend.position = "bottom") +
    #ggplot2::guides(fill = ggplot2::guide_legend(nrow = 3)) +
    ggplot2::theme_minimal() +
    ggplot2::labs( title = chart_title,
                   subtitle = "Counts of patients with comorbidities accross symptoms",
                   y = "Counts",
                   x = "Symptoms",
                   caption = "Source: Dataset - Your.md Dataset") +
    ggplot2::theme(
      axis.title.y = ggplot2::element_text(margin = ggplot2::margin(
        t = 0,
        r = 21,
        b = 0,
        l = 0
      )),
      plot.title = ggplot2::element_text(size = 10, face = "bold"),
      plot.subtitle = ggplot2::element_text(size = 9),
      axis.text.x = ggplot2::element_text(angle = 55, hjust = 1)
    )


plotly::ggplotly(plot_diab_2_num)

II. Association analysis.

Second stage of analysis attempts to analyse whether there are difference in Covid-19 symptom manifestation in different comorbidity groups. As mentioned above, I am stratifying respondents with showing symptoms and those who tested positive. Yet, before we go into actual analysis, correlation of variables is undertaken next.

With the first step being achieved, second step focuses on whether there are differences in Covid-19 symptom in different comorbidity groups. Univariate analysis and multivariate logistic regression has aid the second stage analysis. Before the actual analysis it is worth having a look on how I've stratified the patients for the study.

For the second part, in order to reach the goal of this study, I have stratified patients into two different groups. Patients who showed symptoms but haven't been tested and patients that have been Covid tested positive. The rest of respondents were filtered out of the dataset as this group isn't a relevant group. With the stratification in place, we've got from 60812 respondents, with the stratification of patients, I've got to 9436 respondents. Nevertheless, 8752 are showing symptoms whilst those tested are 707. The groups were recoded into tested and not tested. It is worth mentioning on the class imbalance in this dataset.

data_categ_nosev <- readr::read_csv("/Users/gabrielburcea/Rprojects/data/data_no_sev_stats.csv")

data_categ_nosev <- data_categ_nosev %>%
  dplyr::filter(covid_tested != 'none') %>%
  dplyr::mutate(status_cv = dplyr::case_when(covid_tested == 'showing symptoms' ~ 0,
                                             covid_tested == 'positive' ~ 1)) %>%
  tidyr::drop_na()

show_sympt_positive <- data_categ_nosev %>%
  dplyr::select(id, status_cv) %>%
  dplyr::group_by(status_cv) %>%
  dplyr::tally()

knitr::kable(show_sympt_positive)

How many male and female we have in our dataset? There are 5083 male accounting for 53 percent whilst female are 4321 accounting for 45 percent. Also, we have very low percent of other group.

data_categ_nosev <- readr::read_csv("/Users/gabrielburcea/Rprojects/data/data_no_sev_stats.csv")

gender_tb <- data_categ_nosev %>%
  dplyr::select(ID, gender) %>%
  dplyr::group_by(gender) %>%
  dplyr::tally() %>%
  dplyr::mutate(percent = n/sum(n))

gender_tb

bp <- ggplot2::ggplot(gender_tb, ggplot2::aes(x = "", y = percent, fill = gender)) + 
  ggplot2::geom_bar(width = 1, stat = "identity")

pie <- bp + ggplot2::coord_polar("y", start = 0) + 
  ggplot2::scale_fill_brewer(palette = "Blues") +
  ggplot2::theme(axis.text.x = ggplot2::element_blank())

pie

gender_pie_numbers <- cvindia::gender_distrib(data = data_categ_nosev, start_date = as.Date("2020-04-09"), 
                                                    end_date = as.Date("2020-05-09"), plot_chart = FALSE)

knitr::kable(gender_pie_numbers)

Overview of groups of respondents with symptoms and tested positive.

Looking at respondents who are covid positive and the symptoms.

cov_sympt_plot <- cvindia::covid_positive(data = data_categ_nosev,start_date = as.Date("2020-04-09", tz = "Europe/London"),
                              end_date = as.Date("2020-05-06",tz = "Europe/London"), plot_chart = TRUE)

cov_sympt_plot

cov_sympt_plot <- cvindia::covid_positive(data = data_categ_nosev,start_date = as.Date("2020-04-09", tz = "Europe/London"),
                              end_date = as.Date("2020-05-06",tz = "Europe/London"), plot_chart = FALSE)

knitr::kable(dplyr::top_n(cov_sympt_plot, 10))

Looking at respondents who declared showing symptoms yet no covid test were taken.

show_symp <- cvindia::show_symptoms_cov(data = data_categ_nosev , start_date = as.Date("2020-04-09", tz = "Europe/London"),
                              end_date = as.Date("2020-05-06",tz = "Europe/London"), plot_chart = TRUE)

show_symp

show_symp_num <- cvindia::show_symptoms_cov(data = data_categ_nosev , start_date = as.Date("2020-04-09", tz = "Europe/London"),
                              end_date = as.Date("2020-05-06",tz = "Europe/London"), plot_chart = FALSE)

show_symp_num <- knitr::kable(dplyr::top_n(show_symp_num, 10))

show_symp_num

library(tidymodels)
library(readr)
library(dplyr)
library(corrr)
library(tidyverse)
library(conflicted)
library(tidymodels)
library(ggrepel)
library(corrplot)
library(dplyr)
library(corrr) 
library(themis)
library(rsample)
library(caret)
library(forcats)
library(rcompanion)
library(MASS)
library(pROC)
library(ROCR)
library(data.table)
library(ggplot2)

conflict_prefer("step", "stats")

### ML for Mixed - categorical and numerica data ####

###########################################################################################
### Transforming variables in factor format ###############################################
#data_categ_nosev$country <- as.factor(data_categ_nosev$country)
data_categ_nosev$chills <- as.factor(data_categ_nosev$chills)
data_categ_nosev$cough  <- as.factor(data_categ_nosev$cough)
data_categ_nosev$gender <- as.factor(data_categ_nosev$gender)
data_categ_nosev$covid_tested <- as.factor(data_categ_nosev$covid_tested)
data_categ_nosev$diarrhoea  <- as.factor(data_categ_nosev$diarrhoea)
data_categ_nosev$fatigue  <- as.factor(data_categ_nosev$fatigue)
data_categ_nosev$headache   <- as.factor(data_categ_nosev$headache)
data_categ_nosev$loss_smell_taste   <- as.factor(data_categ_nosev$loss_smell_taste)
data_categ_nosev$muscle_ache  <- as.factor(data_categ_nosev$muscle_ache)
data_categ_nosev$nasal_congestion <- as.factor(data_categ_nosev$nasal_congestion)
data_categ_nosev$nausea_vomiting  <- as.factor(data_categ_nosev$nausea_vomiting)
data_categ_nosev$self_diagnosis <- as.factor(data_categ_nosev$self_diagnosis)
data_categ_nosev$shortness_breath <- as.factor(data_categ_nosev$shortness_breath)
data_categ_nosev$sore_throat <- as.factor(data_categ_nosev$sore_throat)
data_categ_nosev$sputum <- as.factor(data_categ_nosev$sputum)
data_categ_nosev$temperature  <- as.factor(data_categ_nosev$temperature)
data_categ_nosev$health_care_worker <- as.factor(data_categ_nosev$health_care_worker)
data_categ_nosev$care_home_worker <- as.factor(data_categ_nosev$care_home_worker)
### Transforming variables in numerical format  #########################################################
data_categ_nosev$asthma   <- as.factor(data_categ_nosev$asthma)
data_categ_nosev$diabetes_type_two <- as.factor(data_categ_nosev$diabetes_type_two)
data_categ_nosev$obesity <- as.factor(data_categ_nosev$obesity)
data_categ_nosev$hypertension  <- as.factor(data_categ_nosev$hypertension)
data_categ_nosev$heart_disease  <- as.factor(data_categ_nosev$heart_disease)
data_categ_nosev$kidney_disease <- as.factor(data_categ_nosev$kidney_disease)
data_categ_nosev$lung_condition <- as.factor(data_categ_nosev$lung_condition)
data_categ_nosev$liver_disease <- as.factor(data_categ_nosev$liver_disease)
data_categ_nosev$diabetes_type_one <- as.factor(data_categ_nosev$diabetes_type_one)
data_categ_nosev$how_unwell <- as.numeric(data_categ_nosev$how_unwell)
data_categ_nosev$number_days_symptoms <- as.numeric(data_categ_nosev$number_days_symptoms)
data_categ_nosev$age <- as.numeric(data_categ_nosev$age)
data_categ_nosev$age_band <- as.factor(data_categ_nosev$age_band)
data_categ_nosev$temperature <- as.factor(data_categ_nosev$temperature)
#data_categ_nosev$covid_tested <- as.factor(data_categ_nose

Correlations

Correlations matrix is done on numerical variables. However, I have transformed the categorical variables into numerical ones, as I cannot correlate all categorical variables once. Additionally, the correlation is done on original dataset and not on the 9436 datasets since a lot of respondents that declared they have no covid have reported some covid symptoms. However, this has to be discussed more in detail and wait for a bigger dataset.

From the correlation matrix, it is noticed, fatigue and muscle ache seem to be correlated, but only with 0.43550, the highest correlation. Yet, for the highest correlation, it is not even passing the 0.5. However, I choose to take out of the analysis the fatigue variable.

###################################
#######Corelations ################


data_num_with_severity <- read_csv("/Users/gabrielburcea/Rprojects/data/data_numerical.csv")


data_num_with_severity$ID <- NULL
data_num_with_severity$covid_tested <- NULL
data_num_with_severity$country <- NULL
data_num_with_severity$gender <- NULL
data_num_with_severity$age <- NULL

correlation_num_dt <- data_num_with_severity %>% 
dplyr::select_if(is.numeric) %>%
  corrr::correlate(x = .,
                   method = "spearman", 
                   quiet = TRUE) %>%
  corrr::rearrange(x =., 
                   method = "MDS", 
                   absolute = FALSE)

knitr::kable(dplyr::top_n(correlation_num_dt,10))


rplot_labs_v1 <- ggplot2::labs(
  title = "Correlations - symptoms and comorbidities", 
  subtitle = "Symptoms and Commorbidities", 
  caption = "Data accessed from Your.md")

#As we can see, the hue of the color and the size of the point indicate the level of the correlation

correlated_vars_plot <- correlation_num_dt %>%
  corrr::rplot(rdf = ., shape = 19, 
               colors = c("yellow", 
                          "purple")) +
  theme(axis.text.x = element_text(angle = 60, hjust = 1)) +
  rplot_labs_v1

correlated_vars_plot

library(tidymodels)
library(readr)
library(dplyr)
library(corrr)
library(tidyverse)
library(conflicted)
library(tidymodels)
library(ggrepel)
library(corrplot)
library(dplyr)
library(corrr) 
library(themis)
library(rsample)
library(caret)
library(forcats)
library(rcompanion)
library(MASS)
library(pROC)
library(ROCR)
library(data.table)
library(ggplot2)
library(tidyverse)
library(rms)



conflict_prefer("step", "stats")

data_categ_nosev <- readr::read_csv("/Users/gabrielburcea/Rprojects/stats_data_whole/data_categ_no_sev.csv")


level_key_temperature <- 
  c("38.1-39" = "38.2-39",
    "38.1-39" = "Temperature",
    '37' = 'No')

data_categ_nosev <- readr::read_csv("/Users/gabrielburcea/Rprojects/stats_data_whole/data_categ_no_sev.csv")

data_categ_nosev <- data_categ_nosev %>%
  dplyr::filter(covid_tested != 'none') %>%
  dplyr::mutate(status_cv = dplyr::case_when(
                                             covid_tested == 'showing symptoms' ~ 0,
                                             covid_tested == 'positive' ~ 1), 
                number_comorbidities = dplyr::case_when(number_morbidities == 1 ~ 'one', 
                                                        number_morbidities >= 2 ~ 'more than 2'), 
                temperature = forcats::fct_recode(temperature, !!!level_key_temperature)) %>%

  tidyr::drop_na()

data_categ_nosev <- data_categ_nosev %>% dplyr::rename(id = ID)

show_sympt_positive <- data_categ_nosev %>%
  dplyr::select(id, status_cv) %>%
  dplyr::group_by(status_cv) %>%
  dplyr::tally()

knitr::kable(show_sympt_positive)

data_categ_nosev$gender <- as.factor(data_categ_nosev$gender)
#data_categ_nosev$country <- as.factor(data_categ_nosev$country)
data_categ_nosev$chills <- as.factor(data_categ_nosev$chills)
data_categ_nosev$cough  <- as.factor(data_categ_nosev$cough)
data_categ_nosev$diarrhoea  <- as.factor(data_categ_nosev$diarrhoea)
data_categ_nosev$fatigue  <- as.factor(data_categ_nosev$fatigue)
data_categ_nosev$headache   <- as.factor(data_categ_nosev$headache)
data_categ_nosev$loss_smell_taste   <- as.factor(data_categ_nosev$loss_smell_taste)
data_categ_nosev$muscle_ache  <- as.factor(data_categ_nosev$muscle_ache)
data_categ_nosev$nasal_congestion <- as.factor(data_categ_nosev$nasal_congestion)
data_categ_nosev$nausea_vomiting  <- as.factor(data_categ_nosev$nausea_vomiting)
data_categ_nosev$self_diagnosis <- as.factor(data_categ_nosev$self_diagnosis)
data_categ_nosev$shortness_breath <- as.factor(data_categ_nosev$shortness_breath)
data_categ_nosev$sore_throat <- as.factor(data_categ_nosev$sore_throat)
data_categ_nosev$sputum <- as.factor(data_categ_nosev$sputum)
data_categ_nosev$temperature  <- as.factor(data_categ_nosev$temperature)
data_categ_nosev$health_care_worker <- as.factor(data_categ_nosev$health_care_worker)
data_categ_nosev$care_home_worker <- as.factor(data_categ_nosev$care_home_worker)

data_categ_nosev$asthma   <- as.factor(data_categ_nosev$asthma)
data_categ_nosev$diabetes_type_two <- as.factor(data_categ_nosev$diabetes_type_two)
data_categ_nosev$obesity <- as.factor(data_categ_nosev$obesity)
data_categ_nosev$hypertension  <- as.factor(data_categ_nosev$hypertension)
data_categ_nosev$heart_disease  <- as.factor(data_categ_nosev$heart_disease)
data_categ_nosev$kidney_disease <- as.factor(data_categ_nosev$kidney_disease)
data_categ_nosev$lung_condition <- as.factor(data_categ_nosev$lung_condition)
data_categ_nosev$liver_disease <- as.factor(data_categ_nosev$liver_disease)
data_categ_nosev$diabetes_type_one <- as.factor(data_categ_nosev$diabetes_type_one)
data_categ_nosev$how_unwell <- as.factor(data_categ_nosev$how_unwell)
data_categ_nosev$age <- as.factor(data_categ_nosev$age)
data_categ_nosev$covid_tested <- as.factor(data_categ_nosev$covid_tested)

obesity_data <- data_categ_nosev %>%
  dplyr::select(obesity, gender, chills, cough, diarrhoea, headache, loss_smell_taste, muscle_ache, 
                nasal_congestion, nausea_vomiting, shortness_breath, sore_throat, sputum, temperature) %>%
  tidyr::drop_na()

Univariate analysis

Univariate analysis reveal that cough, diarhea, muscle ache, nausea and vomiting, sputum, temperature are associated factors in obese patients.

obesity_chills <- glm(obesity ~ chills, data = obesity_data, family = binomial)

summary(obesity_chills)


coef_obesity_chills <- coef(obesity_chills)

# odd ratios 
odd_ratios_ob_ch <- exp(coef_obesity_chills)
odd_ratios_ob_ch

obesity_cough <- glm(obesity ~ cough, data = obesity_data, family = binomial)

summary(obesity_cough)


coef_obesity_cough <- coef(obesity_cough)

odd_ratios_ob_co <- exp(coef_obesity_cough)

odd_ratios_ob_co

obesity_diarrhea <- glm(obesity ~ diarrhoea, data = obesity_data, family = binomial)

summary(obesity_diarrhea)


# get coef
coef_ob_diarrhea <- coef(obesity_diarrhea)

# odd ratios
odd_ratio_ob_diar <- exp(coef_ob_diarrhea)

odd_ratio_ob_diar

obesity_headache <- glm(obesity ~ headache, data = obesity_data, family = binomial)

summary(obesity_headache)

coef_ob_head <- coef(obesity_headache)

odd_ratio_ob_head <- exp(coef_ob_head)

odd_ratio_ob_head

obesity_loss_smell <- glm(obesity ~ loss_smell_taste, data = obesity_data, family = binomial)


summary(obesity_loss_smell)



coef_ob_loss_smell <- coef(obesity_loss_smell)

odd_ratio_ob_los <- exp(coef_ob_loss_smell)

odd_ratio_ob_los

library(fmsb)

obesity_muscle_ache <- glm(obesity ~ muscle_ache, data = obesity_data, family = binomial)

summary(obesity_muscle_ache)




coef_ob_muscle_ac <- coef(obesity_muscle_ache)

odd_ratio_ob_los <- exp(coef_ob_muscle_ac)

odd_ratio_ob_los

obesity_nasal_cong <- glm(obesity ~ nasal_congestion, data = obesity_data, family = binomial)

summary(obesity_nasal_cong)


coef_ob_nas_cong <- coef(obesity_nasal_cong)

odd_ratio_ob_nas_cong <- exp(coef_ob_nas_cong)

odd_ratio_ob_nas_cong

obesity_nausea_vomitting <- glm(obesity ~ nausea_vomiting, data = obesity_data, family = binomial)

summary(obesity_nausea_vomitting)


coef_ob_naus_vom <- coef(obesity_nausea_vomitting)

odd_ratio_ob_naus_vom <- exp(coef_ob_naus_vom)

odd_ratio_ob_naus_vom

obesity_short_breath <- glm(obesity ~ shortness_breath, data = obesity_data, family = binomial)

summary(obesity_short_breath)

coef_ob_sh_br <- coef(obesity_short_breath)


odd_ratio_ob_sh_br <- exp(coef_ob_sh_br)

odd_ratio_ob_sh_br

obesity_sore_thr <- glm(obesity ~ sore_throat, data = obesity_data, family = binomial)

summary(obesity_sore_thr)

coef_ob_sore_thr <- coef(obesity_sore_thr)


odd_ratio_ob_sore_thr <- exp(coef_ob_sore_thr)

odd_ratio_ob_sore_thr

obesity_sputum <- glm(obesity ~ sputum, data = obesity_data, family = binomial)


summary(obesity_sputum)


coef_ob_sp <- coef(obesity_sputum)

odd_ratio_ob_sp <- exp(coef_ob_sp)

odd_ratio_ob_sp

vif(obesity_sputum)

obesity_temperature <- glm(obesity ~ temperature, data = obesity_data, family = binomial)


summary(obesity_temperature)

coef_ob_temp <- coef(obesity_temperature)

odd_ratio_ob_temp <- exp(coef_ob_temp)

odd_ratio_ob_temp

vif(obesity_temperature)

Adding all variables that showed an association with obesity.

When adjusting for all variables, patients showing covid-19 symptoms/of patients with positive covid test, the results show strong evidence for an association between variables such as cough, diarrhea, sputum and temperature (37.5-38) (p ≤ 0.05) in obese respondents. When adjusting for all variables,in patients showing covid-19 symptpms/of patients with positive covid test, in obese respondents there was: - 15 % increase in odds of experiencing cough compared to those who aren't obese - 35 % increase in diarrhea compared to those who aren't obese - 23 % increase in soutum compared to those who aren't obese - 24 % decrease ??? in 37.5-38 temperature compared to those who do not suffer of heart disease

obesity_model <- glm(obesity ~ cough + diarrhoea + shortness_breath + temperature, data = obesity_data, family = binomial)

summary(obesity_model)


# odds ratio and 95% CL

exp(cbind(OR = coef(obesity_model), confint(obesity_model)))

vif(obesity_model)

asthma_data <- data_categ_nosev %>%
  dplyr::select(asthma, diabetes_type_one, diabetes_type_two, obesity, hypertension, heart_disease, lung_condition, 
                liver_disease, kidney_disease, gender, age, chills, cough, diarrhoea, headache, loss_smell_taste, muscle_ache, 
                nasal_congestion, nausea_vomiting, shortness_breath, sore_throat, sputum, temperature, covid_tested) %>%
  tidyr::drop_na()

asthma_covid <- asthma_data %>%
  dplyr::select(asthma, covid_tested) %>%
  dplyr::group_by(asthma, covid_tested) %>%
  dplyr::tally()

asthma_covid

Univariate analysis for asthma

Unvivariate analysis reveal chills, cough, diarrhea, headache, muschle ache, sore throat, nausea and vomiting, shortness of breath, sputum, temperature are the associated Covid symptoms in respondents with asthma.

Asthma and chills

asthma_chills <- glm(asthma ~ chills, data = asthma_data, family = binomial)

summary(asthma_chills)


coef_asthma_chills <- coef(asthma_chills)

# odd ratios 
odd_ratios_ob_ch <- exp(coef_asthma_chills)
odd_ratios_ob_ch

Asthma and cough

asthma_cough <- glm(asthma ~ cough, data = asthma_data, family = binomial)

summary(asthma_cough)


coef_asthma_cough <- coef(asthma_cough)

odd_ratios_ob_co <- exp(coef_asthma_cough)

odd_ratios_ob_co

Asthma and diarrhea

asthma_diarrhea <- glm(asthma ~ diarrhoea, data = asthma_data, family = binomial)

summary(asthma_diarrhea)


# get coef
coef_ob_diarrhea <- coef(asthma_diarrhea)

# odd ratios
odd_ratio_ob_diar <- exp(coef_ob_diarrhea)

odd_ratio_ob_diar

Asthma and headache

asthma_headache <- glm(asthma ~ headache, data = asthma_data, family = binomial)

summary(asthma_headache)

coef_ob_head <- coef(asthma_headache)

odd_ratio_ob_head <- exp(coef_ob_head)

odd_ratio_ob_head

Asthma and loss of smell and taste

asthma_loss_smell <- glm(asthma ~ loss_smell_taste, data = asthma_data, family = binomial)


summary(asthma_loss_smell)

coef_ob_loss_smell <- coef(asthma_loss_smell)

odd_ratio_ob_los <- (exp(coef_ob_loss_smell)-1)*100

odd_ratio_ob_los

Asthma and muscle ache

asthma_muscle_ache <- glm(asthma ~ muscle_ache, data = asthma_data, family = binomial)

summary(asthma_muscle_ache)

coef_ob_muscle_ac <- coef(asthma_muscle_ache)

odd_ratio_ob_los <- exp(coef_ob_muscle_ac)

odd_ratio_ob_los

Asthma and nasal congestion

asthma_nasal_cong <- glm(asthma ~ nasal_congestion, data = asthma_data, family = binomial)

summary(asthma_nasal_cong)


coef_ob_nas_cong <- coef(asthma_nasal_cong)

odd_ratio_ob_nas_cong <- exp(coef_ob_nas_cong)

odd_ratio_ob_nas_cong

Athma and nausea and vomiting

asthma_nausea_vomitting <- glm(asthma ~ nausea_vomiting, data = asthma_data, family = binomial)

summary(asthma_nausea_vomitting)


coef_ob_naus_vom <- coef(asthma_nausea_vomitting)

odd_ratio_ob_naus_vom <- exp(coef_ob_naus_vom)

odd_ratio_ob_naus_vom

Asthma and shortness of breath

asthma_short_breath <- glm(asthma ~ shortness_breath, data = asthma_data, family = binomial)

summary(asthma_short_breath)

coef_ob_sh_br <- coef(asthma_short_breath)


odd_ratio_ob_sh_br <- exp(coef_ob_sh_br)

odd_ratio_ob_sh_br

Asthma and sore throat

asthma_sore_thr <- glm(asthma ~ sore_throat, data = asthma_data, family = binomial)

summary(asthma_sore_thr)

coef_ob_sore_thr <- coef(asthma_sore_thr)


odd_ratio_ob_sore_thr <- exp(coef_ob_sore_thr)

odd_ratio_ob_sore_thr

Asthma and sputum

asthma_sputum <- glm(asthma ~ sputum, data = asthma_data, family = binomial)


summary(asthma_sputum)


coef_ob_sp <- coef(asthma_sputum)

odd_ratio_ob_sp <- exp(coef_ob_sp)

odd_ratio_ob_sp


vif(asthma_sputum)

Asthma and temperature

asthma_temperature <- glm(asthma ~ temperature, data = asthma_data, family = binomial)

summary(asthma_temperature)

Multivariable Logistic Regression for Asthma

Adding all symptoms that showed to be associated in asthma patients.

When adjusting for all variables,in patients showing covid-19 symptoms/of patients with positive covid test, in patients with asthma there was:

160 % increase in shortness of breath compared to those who do not suffer of asthma
35 % increase in sputum
107 % increase in 39.1-41 temperature compared to those who do not suffer of asthma
1 % increase in male patients compared to those at the baseline
1 % increase in age compared

asthma_model <- glm(asthma ~ chills + cough + diarrhoea + headache + muscle_ache + nausea_vomiting 
                    + shortness_breath + sputum + temperature, data = asthma_data, family = binomial)

summary(asthma_model)

Odd Ratios and 95 % CL for Asthma model

# odds ratios and 95% CL

exp(cbind(OR = coef(asthma_model), confint(asthma_model)))

diabetes_type_one_data <- data_categ_nosev %>%
  dplyr::select(diabetes_type_one, gender,  chills, cough, diarrhoea, headache, loss_smell_taste, muscle_ache, 
                nasal_congestion, nausea_vomiting, shortness_breath, sore_throat, sputum, temperature) %>%
  tidyr::drop_na()

Univariate analysis for diabetes type one

The univariate analysis real that symptoms such as headaches, nasal congestion, sore throat, temperature (39.1-41) are associated with diabetes type one.

diabetes_type_one_chills <- glm(diabetes_type_one ~ chills, data = diabetes_type_one_data, family = binomial)

summary(diabetes_type_one_chills)


coef_diabetes_type_one_chills <- coef(diabetes_type_one_chills)

# odd ratios 
odd_ratios_ob_ch <- exp(coef_diabetes_type_one_chills)
odd_ratios_ob_ch

Diabetes type one with cough

diabetes_type_one_cough <- glm(diabetes_type_one ~ cough, data = diabetes_type_one_data, family = binomial)

summary(diabetes_type_one_cough)


coef_diabetes_type_one_cough <- coef(diabetes_type_one_cough)

odd_ratios_ob_co <- exp(coef_diabetes_type_one_cough)

odd_ratios_ob_co

Diabetes type one with diarrhea

diabetes_type_one_diarrhea <- glm(diabetes_type_one ~ diarrhoea, data = diabetes_type_one_data, family = binomial)

summary(diabetes_type_one_diarrhea)


# get coef
coef_ob_diarrhea <- coef(diabetes_type_one_diarrhea)

# odd ratios
odd_ratio_ob_diar <- exp(coef_ob_diarrhea)

odd_ratio_ob_diar

Diabetes type one with headache

diabetes_type_one_headache <- glm(diabetes_type_one ~ headache, data = diabetes_type_one_data, family = binomial)

summary(diabetes_type_one_headache)

coef_ob_head <- coef(diabetes_type_one_headache)

odd_ratio_ob_head <- exp(coef_ob_head)
odd_ratio_ob_head

Diabetes type one with loss of smell and taste

diabetes_type_one_loss_smell <- glm(diabetes_type_one ~ loss_smell_taste, data = diabetes_type_one_data, family = binomial)


summary(diabetes_type_one_loss_smell)

coef_ob_loss_smell <- coef(diabetes_type_one_loss_smell)

odd_ratio_ob_los <- exp(coef_ob_loss_smell)

odd_ratio_ob_los

Diabetes type one with with muscle ache

diabetes_type_one_muscle_ache <- glm(diabetes_type_one ~ muscle_ache, data = diabetes_type_one_data, family = binomial)

summary(diabetes_type_one_muscle_ache)

coef_ob_muscle_ac <- coef(diabetes_type_one_muscle_ache)

odd_ratio_ob_los <- exp(coef_ob_muscle_ac)

odd_ratio_ob_los

Diabetes type one with nasal congestion

diabetes_type_one_nasal_cong <- glm(diabetes_type_one ~ nasal_congestion, data = diabetes_type_one_data, family = binomial)

summary(diabetes_type_one_nasal_cong)


coef_ob_nas_cong <- coef(diabetes_type_one_nasal_cong)

odd_ratio_ob_nas_cong <- exp(coef_ob_nas_cong)

odd_ratio_ob_nas_cong

Diabetes type one with nausea and vomiting

diabetes_type_one_nausea_vomitting <- glm(diabetes_type_one ~ nausea_vomiting, data = diabetes_type_one_data, family = binomial)

summary(diabetes_type_one_nausea_vomitting)


coef_ob_naus_vom <- coef(diabetes_type_one_nausea_vomitting)

odd_ratio_ob_naus_vom <- exp(coef_ob_naus_vom)

odd_ratio_ob_naus_vom

Diabetest type one with shortness of breath

diabetes_type_one_short_breath <- glm(diabetes_type_one ~ shortness_breath, data = diabetes_type_one_data, family = binomial)

summary(diabetes_type_one_short_breath)

coef_ob_sh_br <- coef(diabetes_type_one_short_breath)


odd_ratio_ob_sh_br <- exp(coef_ob_sh_br)

odd_ratio_ob_sh_br

Diabetes type one with sore throat

diabetes_type_one_sore_thr <- glm(diabetes_type_one ~ sore_throat, data = diabetes_type_one_data, family = binomial)

summary(diabetes_type_one_sore_thr)

coef_ob_sore_thr <- coef(diabetes_type_one_sore_thr)


odd_ratio_ob_sore_thr <- exp(coef_ob_sore_thr)

odd_ratio_ob_sore_thr

Diabetes type one with sputum

diabetes_type_one_sputum <- glm(diabetes_type_one ~ sputum, data = diabetes_type_one_data, family = binomial)


summary(diabetes_type_one_sputum)


coef_ob_sp <- coef(diabetes_type_one_sputum)

odd_ratio_ob_sp <- exp(coef_ob_sp)

odd_ratio_ob_sp

vif(diabetes_type_one_sputum)

Diabetes type one with temperature

diabetes_type_one_temperature <- glm(diabetes_type_one ~ temperature, data = diabetes_type_one_data, family = binomial)


summary(diabetes_type_one_temperature)

coef_ob_temp <- coef(diabetes_type_one_temperature)

odd_ratio_ob_temp <- exp(coef_ob_temp)

odd_ratio_ob_temp

vif(diabetes_type_one_temperature)

Multivariate analysis for diabetest type one

When adding all variables that showed an association with the heart disease model, the model shows something different.

When adjusting for all variables, patients showing covid-19 symptoms/of patients with positive covid test, the results show strong evidence for an association between variables such as headaches and temperature(39.1-41) (p ≤ 0.01) in patients with diabetes type one.
When adjusting for all variables,in patients showing covid-19 symptpms/of patients with positive covid test, in respondent with diabetes type one there was: - 54 % decrease in odds of experiencing headaches compared to those who do not suffer of diabetes type one - 406 % increase in temperature compared to those who do not suffer of heart disease

diabetes_type_one_model <- glm(diabetes_type_one ~ chills + cough + headache + loss_smell_taste + muscle_ache + nausea_vomiting +
                     temperature, data = diabetes_type_one_data, family = binomial)

summary(diabetes_type_one_model)

Odd Ratios and 95 % CL for Diabetes Model

# odds ratio and 95% CL

exp(cbind(OR = coef(diabetes_type_one_model), confint(diabetes_type_one_model)))

vif(diabetes_type_one_model)

diabetes_type_two_data <- data_categ_nosev %>%
  dplyr::select(diabetes_type_two, gender, age, chills, cough, diarrhoea, headache, loss_smell_taste, muscle_ache, 
                nasal_congestion, nausea_vomiting, shortness_breath, sore_throat, sputum, temperature) %>%
  tidyr::drop_na()

Univariate analysis for diabetes type one

Univariate analysis reveal Covid symptoms such as headaches, shortness of breath, sore throat, sputum and temperature are associated wit diabetes type two respondents.

Diabetes type two

diabetes_type_two_chills <- glm(diabetes_type_two ~ chills, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_chills)


coef_diabetes_type_two_chills <- coef(diabetes_type_two_chills)

# odd ratios 
odd_ratios_ob_ch <- exp(coef_diabetes_type_two_chills)
odd_ratios_ob_ch

diabetes_type_two_cough <- glm(diabetes_type_two ~ cough, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_cough)


coef_diabetes_type_two_cough <- coef(diabetes_type_two_cough)

odd_ratios_ob_co <- exp(coef_diabetes_type_two_cough)

odd_ratios_ob_co

diabetes_type_two_diarrhea <- glm(diabetes_type_two ~ diarrhoea, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_diarrhea)


# get coef
coef_ob_diarrhea <- coef(diabetes_type_two_diarrhea)

# odd ratios
odd_ratio_ob_diar <- exp(coef_ob_diarrhea)

odd_ratio_ob_diar

diabetes_type_two_headache <- glm(diabetes_type_two ~ headache, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_headache)

coef_ob_head <- coef(diabetes_type_two_headache)

odd_ratio_ob_head <- exp(coef_ob_head)

odd_ratio_ob_head

diabetes_type_two_loss_smell <- glm(diabetes_type_two ~ loss_smell_taste, data = diabetes_type_two_data, family = binomial)


summary(diabetes_type_two_loss_smell)

coef_ob_loss_smell <- coef(diabetes_type_two_loss_smell)

odd_ratio_ob_los <- exp(coef_ob_loss_smell)

odd_ratio_ob_los

diabetes_type_two_muscle_ache <- glm(diabetes_type_two ~ muscle_ache, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_muscle_ache)

coef_ob_muscle_ac <- coef(diabetes_type_two_muscle_ache)

odd_ratio_ob_los <- exp(coef_ob_muscle_ac)

odd_ratio_ob_los

diabetes_type_two_nasal_cong <- glm(diabetes_type_two ~ nasal_congestion, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_nasal_cong)


coef_ob_nas_cong <- coef(diabetes_type_two_nasal_cong)

odd_ratio_ob_nas_cong <- exp(coef_ob_nas_cong)

odd_ratio_ob_nas_cong

diabetes_type_two_nausea_vomitting <- glm(diabetes_type_two ~ nausea_vomiting, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_nausea_vomitting)


coef_ob_naus_vom <- coef(diabetes_type_two_nausea_vomitting)

odd_ratio_ob_naus_vom <- exp(coef_ob_naus_vom)

odd_ratio_ob_naus_vom

diabetes_type_two_short_breath <- glm(diabetes_type_two ~ shortness_breath, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_short_breath)

coef_ob_sh_br <- coef(diabetes_type_two_short_breath)


odd_ratio_ob_sh_br <- exp(coef_ob_sh_br)

odd_ratio_ob_sh_br

diabetes_type_two_sore_thr <- glm(diabetes_type_two ~ sore_throat, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_sore_thr)

coef_ob_sore_thr <- coef(diabetes_type_two_sore_thr)


odd_ratio_ob_sore_thr <- exp(coef_ob_sore_thr)

odd_ratio_ob_sore_thr

diabetes_type_two_sputum <- glm(diabetes_type_two ~ sputum, data = diabetes_type_two_data, family = binomial)


summary(diabetes_type_two_sputum)


coef_ob_sp <- coef(diabetes_type_two_sputum)

odd_ratio_ob_sp <- exp(coef_ob_sp)

odd_ratio_ob_sp

vif(diabetes_type_two_sputum)

diabetes_type_two_temperature <- glm(diabetes_type_two ~ temperature, data = diabetes_type_two_data, family = binomial)


summary(diabetes_type_two_temperature)

coef_ob_temp <- coef(diabetes_type_two_temperature)

odd_ratio_ob_temp <- exp(coef_ob_temp)

odd_ratio_ob_temp

vif(diabetes_type_two_temperature)

Multivariable analysis for diabetes type two

Adding all variables that showed an association with diabetes type two variable.

When adjusting for all variables, patients showing covid-19 symptoms/of patients with positive covid test, the results show strong evidence for an association between variables such as headache, shortness of breath, sore throat, sputum, temperature age and gender (p ≤ 0.05) in patients with heart disease.
When adjusting for all variables,in patients showing covid-19 symptpms/of patients with positive covid test, in patients with diabetes type two there was: - 26 % decrease in odds of experiencing headaches - 21 % decrease in shortness of breath - 28 % decrease in sputum compared to those who do not suffer of patientss of diabetes type two - 50 % increase in 37.5-38 temperature compared to those who do not suffer of diabetes type two - 87 % increase in 38.1-39 temperature compared to those who do not suffer of diabetes type two - 222 % increase ?? in 39.1-41 temperature - 79 % for gender ?

diabetes_type_two_model <- glm(diabetes_type_two ~ headache + shortness_breath + sore_throat + sputum + temperature, 
                               data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_model)

Odd Ratios and 95 % CL for

exp(cbind(OR = coef(diabetes_type_two_model), confint(diabetes_type_two_model)))

vif(diabetes_type_two_model)

diabetes_type_two_data <- data_categ_nosev %>%
  dplyr::select(diabetes_type_two, gender, age, age_band, chills, cough, diarrhoea, headache, loss_smell_taste, muscle_ache, 
                nasal_congestion, nausea_vomiting, shortness_breath, sore_throat, sputum, temperature) %>%
  tidyr::drop_na()

Univariate analysis for diabetes type one

Univariate analysis reveal Covid symptoms such as headaches, shortness of breath, sore throat, sputum and temperature are associated wit diabetes type two respondents.

Diabetes type two with chills

diabetes_type_two_chills <- glm(diabetes_type_two ~ chills, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_chills)


coef_diabetes_type_two_chills <- coef(diabetes_type_two_chills)

# odd ratios 
odd_ratios_ob_ch <- exp(coef_diabetes_type_two_chills)
odd_ratios_ob_ch

Diabetes type two with cough

diabetes_type_two_cough <- glm(diabetes_type_two ~ cough, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_cough)


coef_diabetes_type_two_cough <- coef(diabetes_type_two_cough)

odd_ratios_ob_co <- exp(coef_diabetes_type_two_cough)

odd_ratios_ob_co

Diabetes type two with diarrhea

diabetes_type_two_diarrhea <- glm(diabetes_type_two ~ diarrhoea, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_diarrhea)


# get coef
coef_ob_diarrhea <- coef(diabetes_type_two_diarrhea)

# odd ratios
odd_ratio_ob_diar <- exp(coef_ob_diarrhea)

odd_ratio_ob_diar

Diabetes type two with headache

diabetes_type_two_headache <- glm(diabetes_type_two ~ headache, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_headache)

coef_ob_head <- coef(diabetes_type_two_headache)

odd_ratio_ob_head <- exp(coef_ob_head)

odd_ratio_ob_head

Diabetes type two with loss of smell and taste

diabetes_type_two_loss_smell <- glm(diabetes_type_two ~ loss_smell_taste, data = diabetes_type_two_data, family = binomial)


summary(diabetes_type_two_loss_smell)

coef_ob_loss_smell <- coef(diabetes_type_two_loss_smell)

odd_ratio_ob_los <- exp(coef_ob_loss_smell)

odd_ratio_ob_los

Diabetes tyoe tow with muscle ache

diabetes_type_two_muscle_ache <- glm(diabetes_type_two ~ muscle_ache, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_muscle_ache)

coef_ob_muscle_ac <- coef(diabetes_type_two_muscle_ache)

odd_ratio_ob_los <- exp(coef_ob_muscle_ac)

odd_ratio_ob_los

Diabetes type two with nasal congestion

diabetes_type_two_nasal_cong <- glm(diabetes_type_two ~ nasal_congestion, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_nasal_cong)


coef_ob_nas_cong <- coef(diabetes_type_two_nasal_cong)

odd_ratio_ob_nas_cong <- exp(coef_ob_nas_cong)

odd_ratio_ob_nas_cong

Diabetes type two with nausea and vomiting

diabetes_type_two_nausea_vomitting <- glm(diabetes_type_two ~ nausea_vomiting, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_nausea_vomitting)


coef_ob_naus_vom <- coef(diabetes_type_two_nausea_vomitting)

odd_ratio_ob_naus_vom <- exp(coef_ob_naus_vom)

odd_ratio_ob_naus_vom

Diabetes type two with shortness of breath

diabetes_type_two_short_breath <- glm(diabetes_type_two ~ shortness_breath, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_short_breath)

coef_ob_sh_br <- coef(diabetes_type_two_short_breath)


odd_ratio_ob_sh_br <- exp(coef_ob_sh_br)

odd_ratio_ob_sh_br

Diabetes type two with sore throat

diabetes_type_two_sore_thr <- glm(diabetes_type_two ~ sore_throat, data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_sore_thr)

coef_ob_sore_thr <- coef(diabetes_type_two_sore_thr)


odd_ratio_ob_sore_thr <- exp(coef_ob_sore_thr)

odd_ratio_ob_sore_thr

Diabetes type two with sputum

diabetes_type_two_sputum <- glm(diabetes_type_two ~ sputum, data = diabetes_type_two_data, family = binomial)


summary(diabetes_type_two_sputum)


coef_ob_sp <- coef(diabetes_type_two_sputum)

odd_ratio_ob_sp <- exp(coef_ob_sp)

odd_ratio_ob_sp

vif(diabetes_type_two_sputum)

Diabetes type two with temperature

diabetes_type_two_temperature <- glm(diabetes_type_two ~ temperature, data = diabetes_type_two_data, family = binomial)


summary(diabetes_type_two_temperature)

coef_ob_temp <- coef(diabetes_type_two_temperature)

odd_ratio_ob_temp <- exp(coef_ob_temp)

odd_ratio_ob_temp

vif(diabetes_type_two_temperature)

Multivariable analysis for diabetes type two

Adding all variables that showed an association with diabetes type two variable.

When adjusting for all variables, patients showing covid-19 symptoms/of patients with positive covid test, the results show strong evidence for an association between variables such as headache, shortness of breath, sore throat, sputum, temperature age and gender (p ≤ 0.05) in patients with heart disease.
When adjusting for all variables,in patients showing covid-19 symptpms/of patients with positive covid test, in patients with diabetes type two there was: - 26 % decrease in odds of experiencing headaches - 21 % decrease in shortness of breath - 28 % decrease in sputum compared to those who do not suffer of patientss of diabetes type two - 50 % increase in 37.5-38 temperature compared to those who do not suffer of diabetes type two - 87 % increase in 38.1-39 temperature compared to those who do not suffer of diabetes type two - 222 % increase ?? in 39.1-41 temperature - 79 % for gender ?

diabetes_type_two_model <- glm(diabetes_type_two ~ headache + shortness_breath + sore_throat + sputum + temperature, 
                               data = diabetes_type_two_data, family = binomial)

summary(diabetes_type_two_model)


exp(cbind(OR = coef(diabetes_type_two_model), confint(diabetes_type_two_model)))
vif(diabetes_type_two_model)

heart_data <- data_categ_nosev %>%
  dplyr::select(heart_disease, gender, age, chills, cough, diarrhoea, headache, loss_smell_taste, muscle_ache, 
                nasal_congestion, nausea_vomiting, shortness_breath, sore_throat, sputum, temperature) %>%
  tidyr::drop_na()

Univariate analysis

Univariate analysis reveal Covid symptoms such as chills, loss of smell and taste, diarhoea, muscle ache , nausea and vomiting, shortness of breath, sputum and temperature(39.1-41) are associated with heart disease.

Heart disease and chills

heart_chills <- glm(heart_disease ~ chills, data = heart_data, family = binomial)

summary(heart_chills)


coef_heart_chills <- coef(heart_chills)

# odd ratios 
odd_ratios_ob_ch <- exp(coef_heart_chills)
odd_ratios_ob_ch

Heart disease and cough

heart_cough <- glm(heart_disease ~ cough, data = heart_data, family = binomial)

summary(heart_cough)


coef_heart_cough <- coef(heart_cough)

odd_ratios_ob_co <- exp(coef_heart_cough)

odd_ratios_ob_co

Heart disease and diarrhea

heart_diarrhea <- glm(heart_disease ~ diarrhoea, data = heart_data, family = binomial)

summary(heart_diarrhea)


# get coef
coef_ob_diarrhea <- coef(heart_diarrhea)

# odd ratios
odd_ratio_ob_diar <- exp(coef_ob_diarrhea)

odd_ratio_ob_diar

Heart disease and headache

heart_headache <- glm(heart_disease ~ headache, data = heart_data, family = binomial)

summary(heart_headache)

coef_ob_head <- coef(heart_headache)

odd_ratio_ob_head <- exp(coef_ob_head)

odd_ratio_ob_head

Heart disease and loss of smell and taste

heart_loss_smell <- glm(heart_disease ~ loss_smell_taste, data = heart_data, family = binomial)


summary(heart_loss_smell)

coef_ob_loss_smell <- coef(heart_loss_smell)

odd_ratio_ob_los <- exp(coef_ob_loss_smell)

odd_ratio_ob_los

Heart disease and muscle ache.

heart_muscle_ache <- glm(heart_disease ~ muscle_ache, data = heart_data, family = binomial)

summary(heart_muscle_ache)

coef_ob_muscle_ac <- coef(heart_muscle_ache)

odd_ratio_ob_los <- exp(coef_ob_muscle_ac)

odd_ratio_ob_los

Heart disease and nasal congestion

heart_nasal_cong <- glm(heart_disease ~ nasal_congestion, data = heart_data, family = binomial)

summary(heart_nasal_cong)


coef_ob_nas_cong <- coef(heart_nasal_cong)

odd_ratio_ob_nas_cong <- exp(coef_ob_nas_cong)

odd_ratio_ob_nas_cong

Heart disease and nausea and vomiting

heart_nausea_vomitting <- glm(heart_disease ~ nausea_vomiting, data = heart_data, family = binomial)

summary(heart_nausea_vomitting)


coef_ob_naus_vom <- coef(heart_nausea_vomitting)

odd_ratio_ob_naus_vom <- exp(coef_ob_naus_vom)

odd_ratio_ob_naus_vom

Heart disease and shortness of breath

heart_short_breath <- glm(heart_disease ~ shortness_breath, data = heart_data, family = binomial)

summary(heart_short_breath)

coef_ob_sh_br <- coef(heart_short_breath)


odd_ratio_ob_sh_br <- exp(coef_ob_sh_br)

odd_ratio_ob_sh_br

Heart disease and sore throat

heart_sore_thr <- glm(heart_disease ~ sore_throat, data = heart_data, family = binomial)

summary(heart_sore_thr)

coef_ob_sore_thr <- coef(heart_sore_thr)


odd_ratio_ob_sore_thr <- exp(coef_ob_sore_thr)

odd_ratio_ob_sore_thr

11.Heart disease and sputum

heart_sputum <- glm(heart_disease ~ sputum, data = heart_data, family = binomial)


summary(heart_sputum)


coef_ob_sp <- coef(heart_sputum)

odd_ratio_ob_sp <- exp(coef_ob_sp)

odd_ratio_ob_sp

vif(heart_sputum)

Heart disease and temperature

heart_temperature <- glm(heart_disease ~ temperature, data = heart_data, family = binomial)


summary(heart_temperature)

coef_ob_temp <- coef(heart_temperature)

odd_ratio_ob_temp <- exp(coef_ob_temp)

odd_ratio_ob_temp

vif(heart_temperature)

Multivariate analysis for heart disease

Adding all variables that showed an association with the heart disease model.

When adjusting for all variables,in patients showing covid-19 symptpms/of patients with positive covid test, in patients with heart disease there was: - 94 % increase in odds of experiencing chills compared to those who do not suffer of heart diesease disease - 95 % increase in nausea and vomiting compared to those who do not suffer of heart disease - 44 % increase in shortness of breath compared to those who do not suffer of heart disease - 7 % increase in 37.5-38 temperature compared to those who do not suffer of heart disease - 229 % increase in 39.1-41 temperature

heart_model <- glm(heart_disease ~ chills + diarrhoea + loss_smell_taste +muscle_ache + nausea_vomiting + shortness_breath +
                     temperature + sputum  + shortness_breath, data = heart_data, family = binomial)

summary(heart_model)

exp(cbind(OR = coef(heart_model), confint(heart_model)))

vif(heart_model)

hypertension_data <- data_categ_nosev %>%
  dplyr::select(hypertension, gender, age, age_band, chills, cough, diarrhoea, headache, loss_smell_taste, muscle_ache, 
                nasal_congestion, nausea_vomiting, shortness_breath, sore_throat, sputum, temperature) %>%
  tidyr::drop_na()

Univariate analysis for hypertension and Covid symptoms

Univariate analysis reveal cough, diarrhea, muscle ache, shortness of breath, sputum and temperature (38.1-39) are symptoms associated in patients with hypertension.

1.Hypertension and chills

hypertension_chills <- glm(hypertension ~ chills, data = hypertension_data, family = binomial)

summary(hypertension_chills)


coef_hypertension_chills <- coef(hypertension_chills)

# odd ratios 
odd_ratios_ob_ch <- exp(coef_hypertension_chills)
odd_ratios_ob_ch

Hypertension and cough

hypertension_cough <- glm(hypertension ~ cough, data = hypertension_data, family = binomial)

summary(hypertension_cough)


coef_hypertension_cough <- coef(hypertension_cough)

odd_ratios_ob_co <- exp(coef_hypertension_cough)

odd_ratios_ob_co

Hypertension and diarrhea

hypertension_diarrhea <- glm(hypertension ~ diarrhoea, data = hypertension_data, family = binomial)

summary(hypertension_diarrhea)


# get coef
coef_ob_diarrhea <- coef(hypertension_diarrhea)

# odd ratios
odd_ratio_ob_diar <- exp(coef_ob_diarrhea)

odd_ratio_ob_diar

Hypertension and headache

hypertension_headache <- glm(hypertension ~ headache, data = hypertension_data, family = binomial)

summary(hypertension_headache)

coef_ob_head <- coef(hypertension_headache)

odd_ratio_ob_head <- exp(coef_ob_head)

odd_ratio_ob_head

Hypertension and loss of smell and taste

hypertension_loss_smell <- glm(hypertension ~ loss_smell_taste, data = hypertension_data, family = binomial)


summary(hypertension_loss_smell)

coef_ob_loss_smell <- coef(hypertension_loss_smell)

odd_ratio_ob_los <- exp(coef_ob_loss_smell)

odd_ratio_ob_los

Hypertension and muscle ache

hypertension_muscle_ache <- glm(hypertension ~ muscle_ache, data = hypertension_data, family = binomial)

summary(hypertension_muscle_ache)

coef_ob_muscle_ac <- coef(hypertension_muscle_ache)

odd_ratio_ob_los <- exp(coef_ob_muscle_ac)

odd_ratio_ob_los

Hypertension and nasal congestion

hypertension_nasal_cong <- glm(hypertension ~ nasal_congestion, data = hypertension_data, family = binomial)

summary(hypertension_nasal_cong)


coef_ob_nas_cong <- coef(hypertension_nasal_cong)

odd_ratio_ob_nas_cong <- exp(coef_ob_nas_cong)

odd_ratio_ob_nas_cong

Hypertension and nausea and vomiting

hypertension_nausea_vomitting <- glm(hypertension ~ nausea_vomiting, data = hypertension_data, family = binomial)

summary(hypertension_nausea_vomitting)


coef_ob_naus_vom <- coef(hypertension_nausea_vomitting)

odd_ratio_ob_naus_vom <- exp(coef_ob_naus_vom)

odd_ratio_ob_naus_vom

Hypertension and shortness of breath

hypertension_short_breath <- glm(hypertension ~ shortness_breath, data = hypertension_data, family = binomial)

summary(hypertension_short_breath)

coef_ob_sh_br <- coef(hypertension_short_breath)


odd_ratio_ob_sh_br <- exp(coef_ob_sh_br)

odd_ratio_ob_sh_br

Hypertension and sore throat

hypertension_sore_thr <- glm(hypertension ~ sore_throat, data = hypertension_data, family = binomial)

summary(hypertension_sore_thr)

coef_ob_sore_thr <- coef(hypertension_sore_thr)


odd_ratio_ob_sore_thr <- exp(coef_ob_sore_thr)

odd_ratio_ob_sore_thr

Hypertension and sputum

hypertension_sputum <- glm(hypertension ~ sputum, data = hypertension_data, family = binomial)


summary(hypertension_sputum)


coef_ob_sp <- coef(hypertension_sputum)

odd_ratio_ob_sp <- exp(coef_ob_sp)

odd_ratio_ob_sp

vif(hypertension_sputum)

Hypertension and temperature

hypertension_temperature <- glm(hypertension ~ temperature, data = hypertension_data, family = binomial)


summary(hypertension_temperature)

coef_ob_temp <- coef(hypertension_temperature)

odd_ratio_ob_temp <- exp(coef_ob_temp)

odd_ratio_ob_temp

vif(hypertension_temperature)

Multivariable analysis for hypertension

Adding all variables that showed an association with the hypertension variable.

When adjusting for all variables, patients showing covid-19 symptoms/of patients with positive covid test, the results show strong evidence for an association between variables such as cough, muscle ache, shortness of breath, sputum and temperature (37.1- 38 ; 38.1-39) (p ≤ 0.05) in patients with hypertension.
When adjusting for all variables,in patients showing covid-19 symptpms/of patients with positive covid test, in patients with hypertension there was: - 24 % increase in odds of experiencing cough compared to those who do not suffer of hypertension - 15 % increase in diarrhea compared to those who do not suffer of hypertension - 12 % increase in muscle ache compared to those who do not suffer of hypertension - 5 % increase in sputum compared to those who do not suffer of hypertension - 51 % increase in 38.1-39 in temperature to those who do not suffer of hypertension - 21 % increase in 39.1-41 in temperature compared to those who do not suffer of hypertension

cough, diarrhea, muscle ache, shortness of breath, sputum, temperature

hypertension_model <- glm(hypertension ~ cough + diarrhoea + muscle_ache + shortness_breath + sputum + temperature,  data = hypertension_data, family = binomial)

summary(hypertension_model)

exp(cbind(OR = coef(hypertension_model), confint(hypertension_model)))

vif(hypertension_model)

liver_data <- data_categ_nosev %>%
  dplyr::select(liver_disease, gender, age, chills, cough, diarrhoea, headache, loss_smell_taste, muscle_ache, 
                nasal_congestion, nausea_vomiting, shortness_breath, sore_throat, sputum, temperature) %>%
  tidyr::drop_na()

Univariate analysis for liver disease and Covid-19 symptoms

Asociated Covid symptoms in respondents with liver disease are: chills, diarhoea, loss of smell and taste, muscle ache, nasal congestion, nausea and vomiting, sputum and temperature

Liver disease and chills

liver_chills <- glm(liver_disease ~ chills, data = liver_data, family = binomial)

summary(liver_chills)


coef_liver_chills <- coef(liver_chills)

# odd ratios 
odd_ratios_ob_ch <- exp(coef_liver_chills)
odd_ratios_ob_ch

Liver disease and cough

liver_cough <- glm(liver_disease ~ cough, data = liver_data, family = binomial)

summary(liver_cough)


coef_liver_cough <- coef(liver_cough)

odd_ratios_ob_co <- exp(coef_liver_cough)

odd_ratios_ob_co

Liver disease and diarhoea

liver_diarrhea <- glm(liver_disease ~ diarrhoea, data = liver_data, family = binomial)

summary(liver_diarrhea)


# get coef
coef_ob_diarrhea <- coef(liver_diarrhea)

# odd ratios
odd_ratio_ob_diar <- exp(coef_ob_diarrhea)

odd_ratio_ob_diar

Liver disease and headache

liver_headache <- glm(liver_disease ~ headache, data = liver_data, family = binomial)

summary(liver_headache)

coef_ob_head <- coef(liver_headache)

odd_ratio_ob_head <- exp(coef_ob_head)

odd_ratio_ob_head

Liver disease and loss of smell and taste

liver_loss_smell <- glm(liver_disease ~ loss_smell_taste, data = liver_data, family = binomial)


summary(liver_loss_smell)

coef_ob_loss_smell <- coef(liver_loss_smell)

odd_ratio_ob_los <- exp(coef_ob_loss_smell)

odd_ratio_ob_los

Liver disease and muscle ache

liver_muscle_ache <- glm(liver_disease ~ muscle_ache, data = liver_data, family = binomial)

summary(liver_muscle_ache)

coef_ob_muscle_ac <- coef(liver_muscle_ache)

odd_ratio_ob_los <- exp(coef_ob_muscle_ac)

odd_ratio_ob_los

Liver disease and nasal congestion

liver_nasal_cong <- glm(liver_disease ~ nasal_congestion, data = liver_data, family = binomial)

summary(liver_nasal_cong)


coef_ob_nas_cong <- coef(liver_nasal_cong)

odd_ratio_ob_nas_cong <- exp(coef_ob_nas_cong)

odd_ratio_ob_nas_cong

Liver disease and vomiting

liver_nausea_vomitting <- glm(liver_disease ~ nausea_vomiting, data = liver_data, family = binomial)

summary(liver_nausea_vomitting)


coef_ob_naus_vom <- coef(liver_nausea_vomitting)

odd_ratio_ob_naus_vom <- exp(coef_ob_naus_vom)

odd_ratio_ob_naus_vom

Liver disease and shortness of breath

liver_short_breath <- glm(liver_disease ~ shortness_breath, data = liver_data, family = binomial)

summary(liver_short_breath)

coef_ob_sh_br <- coef(liver_short_breath)


odd_ratio_ob_sh_br <- exp(coef_ob_sh_br)

odd_ratio_ob_sh_br

Liver disease and sore throat

liver_sore_thr <- glm(liver_disease ~ sore_throat, data = liver_data, family = binomial)

summary(liver_sore_thr)

coef_ob_sore_thr <- coef(liver_sore_thr)


odd_ratio_ob_sore_thr <- exp(coef_ob_sore_thr)

odd_ratio_ob_sore_thr

Liver disease and sputum

liver_sputum <- glm(liver_disease ~ sputum, data = liver_data, family = binomial)


summary(liver_sputum)


coef_ob_sp <- coef(liver_sputum)

odd_ratio_ob_sp <- exp(coef_ob_sp)

odd_ratio_ob_sp

vif(liver_sputum)

Liver disease and temperature

liver_temperature <- glm(liver_disease ~ temperature, data = liver_data, family = binomial)


summary(liver_temperature)

coef_ob_temp <- coef(liver_temperature)

odd_ratio_ob_temp <- exp(coef_ob_temp)

odd_ratio_ob_temp

vif(liver_temperature)

Multivariate analysis for liver disease

When adjusting for the variables, in patients showing covid-19 symptoms/of patients with positive covid test the results show strong evidence for an association between variables such as chills, nausea and vomiting, and temperature (38.1-39; 39.1-40) (p ≤ 0.05) for patients with liver disease.
When adjusting for all variables,in patients showing covid-19 symptpms/of patients with positive covid test, in patients with liver disease there was: - 103 % increase in odds of experiencing chills compared to those who do not suffer of liver disease - 13 % increase in loss of smell and taste compared to those who do not suffer of liver disease - 15 % increase in muscle ache compred to those who do not suffer of liver disease - 272 % increase in nausea and vomiting compared to those who do not suffer of liver disease -17 % increase in nasal congestion compared to those who do not suffer of liver disease - 14 % increase in 37.5-38 temperature compared to those who do not suffer of liver disease - 98 % increase in 38.1-39 temperature compared to those who do not suffer of liver disease - 338 % increase ?? in 39.1-41 temperature - 21 % increase in sputum compared to those who do not suffer of liver disease

liver_model <- glm(liver_disease ~ chills + diarrhoea + loss_smell_taste + muscle_ache + nausea_vomiting + nasal_congestion + shortness_breath +
                     temperature + sputum, data = liver_data, family = binomial)

summary(liver_model)

exp(cbind(OR = coef(liver_model), confint(liver_model)))

vif(liver_model)

lung_condition_data <- data_categ_nosev %>%
  dplyr::select(lung_condition, gender, age, chills, cough, diarrhoea, headache, loss_smell_taste, muscle_ache, 
                nasal_congestion, nausea_vomiting, shortness_breath, sore_throat, sputum, temperature) %>%
  tidyr::drop_na()

Univariate analysis for lung condition and Covid-19 symptoms

Univariate logistic regression for lung condition and covid symptoms reveal: - nausea and vomiting, shortness of breath, sputum and 39.1- 41 temperature are statistically significant by looking at p values

Lung condition and chills

lung_condition_chills <- glm(lung_condition ~ chills, data = lung_condition_data, family = binomial)

summary(lung_condition_chills)


coef_lung_condition_chills <- coef(lung_condition_chills)

# odd ratios 
odd_ratios_ob_ch <- exp(coef_lung_condition_chills)
odd_ratios_ob_ch

Lung condition and cough

lung_condition_cough <- glm(lung_condition ~ cough, data = lung_condition_data, family = binomial)

summary(lung_condition_cough)


coef_lung_condition_cough <- coef(lung_condition_cough)

odd_ratios_ob_co <- exp(coef_lung_condition_cough)

odd_ratios_ob_co

Lung condition and diarhoea

lung_condition_diarrhea <- glm(lung_condition ~ diarrhoea, data = lung_condition_data, family = binomial)

summary(lung_condition_diarrhea)


# get coef
coef_ob_diarrhea <- coef(lung_condition_diarrhea)

# odd ratios
odd_ratio_ob_diar <- exp(coef_ob_diarrhea)

odd_ratio_ob_diar

Lung condition and headache

lung_condition_headache <- glm(lung_condition ~ headache, data = lung_condition_data, family = binomial)

summary(lung_condition_headache)

coef_ob_head <- coef(lung_condition_headache)

odd_ratio_ob_head <- exp(coef_ob_head)

odd_ratio_ob_head

Lung condition and loss of smell and taste

lung_condition_loss_smell <- glm(lung_condition ~ loss_smell_taste, data = lung_condition_data, family = binomial)


summary(lung_condition_loss_smell)

coef_ob_loss_smell <- coef(lung_condition_loss_smell)

odd_ratio_ob_los <- exp(coef_ob_loss_smell)

odd_ratio_ob_los

Lung condition and muscle ache

lung_condition_muscle_ache <- glm(lung_condition ~ muscle_ache, data = lung_condition_data, family = binomial)

summary(lung_condition_muscle_ache)

coef_ob_muscle_ac <- coef(lung_condition_muscle_ache)

odd_ratio_ob_los <- exp(coef_ob_muscle_ac)

odd_ratio_ob_los

Lung condition and nasal congestion

lung_condition_nasal_cong <- glm(lung_condition ~ nasal_congestion, data = lung_condition_data, family = binomial)

summary(lung_condition_nasal_cong)


coef_ob_nas_cong <- coef(lung_condition_nasal_cong)

odd_ratio_ob_nas_cong <- exp(coef_ob_nas_cong)

odd_ratio_ob_nas_cong

Lung condition and nausea and vomiting

lung_condition_nausea_vomitting <- glm(lung_condition ~ nausea_vomiting, data = lung_condition_data, family = binomial)

summary(lung_condition_nausea_vomitting)


coef_ob_naus_vom <- coef(lung_condition_nausea_vomitting)

odd_ratio_ob_naus_vom <- exp(coef_ob_naus_vom)

odd_ratio_ob_naus_vom

Lung condition and shorthness of breath

lung_condition_short_breath <- glm(lung_condition ~ shortness_breath, data = lung_condition_data, family = binomial)

summary(lung_condition_short_breath)

coef_ob_sh_br <- coef(lung_condition_short_breath)


odd_ratio_ob_sh_br <- exp(coef_ob_sh_br)

odd_ratio_ob_sh_br

Lung condition and sore throat

lung_condition_sore_thr <- glm(lung_condition ~ sore_throat, data = lung_condition_data, family = binomial)

summary(lung_condition_sore_thr)

coef_ob_sore_thr <- coef(lung_condition_sore_thr)


odd_ratio_ob_sore_thr <- exp(coef_ob_sore_thr)

odd_ratio_ob_sore_thr

Lung condition and sputum

lung_condition_sputum <- glm(lung_condition ~ sputum, data = lung_condition_data, family = binomial)


summary(lung_condition_sputum)


coef_ob_sp <- coef(lung_condition_sputum)

odd_ratio_ob_sp <- exp(coef_ob_sp)

odd_ratio_ob_sp

vif(lung_condition_sputum)

Lung condition and temperature

lung_condition_temperature <- glm(lung_condition ~ temperature, data = lung_condition_data, family = binomial)


summary(lung_condition_temperature)

coef_ob_temp <- coef(lung_condition_temperature)

odd_ratio_ob_temp <- exp(coef_ob_temp)

odd_ratio_ob_temp

vif(lung_condition_temperature)

Multivariate analysis for lung condition

Adding all variables that showed an association with the lung diseease.

When adjusting for the above variables, in patients showing covid-19 symptoms/of patients with positivie covid test, there was: - 88 % increase in the odds of experiencing nausea and vomiting compared to those who did not have lung disease; - 63 % increase in odds of experiencing shortness of breath compared to those who did not have lung disease; - 146 % increase in 39.1-41 temperature compared to those who did not have lung disease; - 33 % increase in the odds of experiencing sputum in patients with lung disease compared to those who did not have lung disease

lung_condition_model <- glm(lung_condition ~  nausea_vomiting +  shortness_breath +
                     temperature + sputum, data = lung_condition_data, family = binomial)

summary(lung_condition_model)

exp(cbind(OR = coef(lung_condition_model), confint(lung_condition_model)))


vif(lung_condition_model)

kidney_data <- data_categ_nosev %>%
  dplyr::select(kidney_disease, gender, age,  chills, cough, diarrhoea, headache, loss_smell_taste, muscle_ache, 
                nasal_congestion, nausea_vomiting, shortness_breath, sore_throat, sputum, temperature) %>%
  tidyr::drop_na()

Univariate analysis for kidney disease and Covid-19 symptoms

The most Covid symptoms that are associated with kidney disease are shorthness and breath and temperature.

Kidney disease and chills

kidney_chills <- glm(kidney_disease ~ chills, data = kidney_data, family = binomial)

summary(kidney_chills)


coef_kidney_chills <- coef(kidney_chills)

# odd ratios 
odd_ratios_ob_ch <- exp(coef_kidney_chills)
odd_ratios_ob_ch

Kidney disease and cough

kidney_cough <- glm(kidney_disease ~ cough, data = kidney_data, family = binomial)

summary(kidney_cough)


coef_kidney_cough <- coef(kidney_cough)

odd_ratios_ob_co <- exp(coef_kidney_cough)

odd_ratios_ob_co

Kidney disease and diarrhoea

kidney_diarrhea <- glm(kidney_disease ~ diarrhoea, data = kidney_data, family = binomial)

summary(kidney_diarrhea)


# get coef
coef_ob_diarrhea <- coef(kidney_diarrhea)

# odd ratios
odd_ratio_ob_diar <- exp(coef_ob_diarrhea)

odd_ratio_ob_diar

Kidney disease and headache

kidney_headache <- glm(kidney_disease ~ headache, data = kidney_data, family = binomial)

summary(kidney_headache)

coef_ob_head <- coef(kidney_headache)

odd_ratio_ob_head <- exp(coef_ob_head)

odd_ratio_ob_head

Kidney disease and loss of smell and taste

kidney_loss_smell <- glm(kidney_disease ~ loss_smell_taste, data = kidney_data, family = binomial)


summary(kidney_loss_smell)

coef_ob_loss_smell <- coef(kidney_loss_smell)

odd_ratio_ob_los <- exp(coef_ob_loss_smell)

odd_ratio_ob_los

Kidney disease and muscle ache

kidney_muscle_ache <- glm(kidney_disease ~ muscle_ache, data = kidney_data, family = binomial)

summary(kidney_muscle_ache)

coef_ob_muscle_ac <- coef(kidney_muscle_ache)

odd_ratio_ob_los <- exp(coef_ob_muscle_ac)

odd_ratio_ob_los

Kidney disease and nasal congestion

kidney_nasal_cong <- glm(kidney_disease ~ nasal_congestion, data = kidney_data, family = binomial)

summary(kidney_nasal_cong)


coef_ob_nas_cong <- coef(kidney_nasal_cong)

odd_ratio_ob_nas_cong <- exp(coef_ob_nas_cong)

odd_ratio_ob_nas_cong

Kidney disease and nausea and vomiting

kidney_nausea_vomitting <- glm(kidney_disease ~ nausea_vomiting, data = kidney_data, family = binomial)

summary(kidney_nausea_vomitting)


coef_ob_naus_vom <- coef(kidney_nausea_vomitting)

odd_ratio_ob_naus_vom <- exp(coef_ob_naus_vom)

odd_ratio_ob_naus_vom

Kidney disease and shorthness of breath

kidney_short_breath <- glm(kidney_disease ~ shortness_breath, data = kidney_data, family = binomial)

summary(kidney_short_breath)

coef_ob_sh_br <- coef(kidney_short_breath)


odd_ratio_ob_sh_br <- exp(coef_ob_sh_br)

odd_ratio_ob_sh_br

Kidney disease and sore throat

kidney_sore_thr <- glm(kidney_disease ~ sore_throat, data = kidney_data, family = binomial)

summary(kidney_sore_thr)

coef_ob_sore_thr <- coef(kidney_sore_thr)


odd_ratio_ob_sore_thr <- exp(coef_ob_sore_thr)

odd_ratio_ob_sore_thr

Kidney disease and sputum

kidney_sputum <- glm(kidney_disease ~ sputum, data = kidney_data, family = binomial)


summary(kidney_sputum)


coef_ob_sp <- coef(kidney_sputum)

odd_ratio_ob_sp <- exp(coef_ob_sp)

odd_ratio_ob_sp

vif(kidney_sputum)

Kidney disease and temperature

kidney_temperature <- glm(kidney_disease ~ temperature, data = kidney_data, family = binomial)


summary(kidney_temperature)

coef_ob_temp <- coef(kidney_temperature)

odd_ratio_ob_temp <- exp(coef_ob_temp)

odd_ratio_ob_temp

vif(kidney_temperature)

Multivariate analysis for Kidney disease

The selected variables for the final model are: shortness of breath, temperature.

In patients showing covid-19 symptoms/of patients with a positive covid test,when adjusting for all relevant variables, in respondents with kidney disease there was a: - 64% increase in the odds of experiencing shortness of breaht compared to those who do not have kidney disease - 33 % increase in the odds of experiencing 38.1 -39 temperature compared to those who do not have kidney disease - 305 % ?? not sure - increase in odds of experiencing 39.1-41 temperature compared to those who do not have kidney disease

kidney_model <- glm(kidney_disease ~  shortness_breath + temperature, data = kidney_data, family = binomial)

summary(kidney_model)

exp(cbind(OR = coef(kidney_model), confint(kidney_model)))
vif(kidney_model)

gabrielburcea/cvindia documentation built on Feb. 17, 2021, 3:31 a.m.

rdrr.io home R language documentation Run R code online

CRAN packages Bioconductor packages R-Forge packages GitHub packages

Note that we can't provide technical support on individual packages. You should contact the package authors for that.

gabrielburcea/cvindia
Analysis of symptoms

In gabrielburcea/cvindia: Analysis of symptoms

I. Descriptive analytics

II. Association analysis.

Correlations

Univariate analysis

Univariate analysis for asthma

Multivariable Logistic Regression for Asthma

Univariate analysis for diabetes type one

Multivariate analysis for diabetest type one

Univariate analysis for diabetes type one

Multivariable analysis for diabetes type two

Univariate analysis for diabetes type one

Multivariable analysis for diabetes type two

Univariate analysis

Multivariate analysis for heart disease

Univariate analysis for hypertension and Covid symptoms

Multivariable analysis for hypertension

Univariate analysis for liver disease and Covid-19 symptoms

Multivariate analysis for liver disease

Univariate analysis for lung condition and Covid-19 symptoms

Multivariate analysis for lung condition

Univariate analysis for kidney disease and Covid-19 symptoms

Multivariate analysis for Kidney disease

R Package Documentation

Browse R Packages

We want your feedback!

gabrielburcea/cvindia Analysis of symptoms

In gabrielburcea/cvindia: Analysis of symptoms

I. Descriptive analytics

II. Association analysis.

Correlations

Univariate analysis

Univariate analysis for asthma

Multivariable Logistic Regression for Asthma

Univariate analysis for diabetes type one

Multivariate analysis for diabetest type one

Univariate analysis for diabetes type one

Multivariable analysis for diabetes type two

Univariate analysis for diabetes type one

Multivariable analysis for diabetes type two

Univariate analysis

Multivariate analysis for heart disease

Univariate analysis for hypertension and Covid symptoms

Multivariable analysis for hypertension

Univariate analysis for liver disease and Covid-19 symptoms

Multivariate analysis for liver disease

Univariate analysis for lung condition and Covid-19 symptoms

Multivariate analysis for lung condition

Univariate analysis for kidney disease and Covid-19 symptoms

Multivariate analysis for Kidney disease

R Package Documentation

Browse R Packages

We want your feedback!

gabrielburcea/cvindia
Analysis of symptoms